JP2002342296A - Network system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problem of service level deteriorating, since it is decided that service is active, even though the service is operating, but is hindered in a considerable high loaded state. SOLUTION: The service levels of a plurality of services are measured periodically, and the measured level of each service is compared with a preliminarily set threshold. Also, when the measured level exceeds the threshold based on the compared result, a server for backup is assigned to the service whose measured level exceeds the threshold, so that the server can be reinforced, and deterioration of the service is prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a network system for providing various services on a network.

[0002]

2. Description of the Related Art Conventionally, in such a network system, a plurality of servers are connected on a network (such as an Internet network), and various services are provided according to access from a user. In a conventional system, a method of monitoring the state of a server or a service and switching the server when the server has failed or the service has failed, for example, is used to maintain the service provision.

[0003]

However, in the conventional system, the service is running, but even if the service is interrupted under a considerably high load condition,
Since the service is determined to be alive, there are problems such as a decrease in service level and a delay in service response. In addition, when it is determined that a server or service has gone down, the server is switched to a server that provides a predetermined service as a substitute for that server. Therefore, it is necessary to prepare an alternative server for each service. It became large and inefficient.

[0004] The present invention has been made in view of the above-mentioned conventional problems, and an object of the present invention is to reinforce a server in a timely manner to a service whose service level is decreasing,
An object of the present invention is to provide a network system capable of preventing a service from being deteriorated.

[0005]

In order to achieve the above object, a network system according to the present invention has a network system in which a server is connected to a network corresponding to each of a plurality of services and a plurality of services are provided on the network. Means for periodically measuring service levels of a plurality of services, means for comparing the measured level of each service measured by the measuring means with a preset threshold value, and a comparison result of the comparing means And a means for allocating a backup server to a service whose measurement level exceeds the threshold value when the measurement level exceeds the threshold value and reinforcing the server.

In the present invention, the service level of a plurality of services is periodically measured, and a backup server is allocated to a service whose service level is decreasing to reinforce the server. It can be performed in real time, and even if the load is concentrated on a specific service, the load can be distributed and the service can be prevented from lowering.

[0007]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the first embodiment of the network system of the present invention. In FIG. 1, 101 and 102 are service A
Are provided, and 103 and 104 are servers that provide the service B. W for the servers 101 and 102
It is assumed that the eb server and the servers 103 and 104 are Ftp servers. The service A is, for example, an image display for online shopping, and the service B is, for example, a software download service.

Reference numeral 105 denotes a backup server, which can operate as a Web server or an Ftp server in response to an external request. 106
Is a server that executes a service level monitoring tool.
This service level monitoring tool periodically provides HTTP,
The Internet service level of response time such as FTP is measured, and the measured level is compared with a threshold. Further, based on the comparison result, when the service level such as the response time exceeds the threshold value, the load on the server providing the service becomes large, and it is detected that the response is delayed, and the service level is lowered. The load is distributed by allocating the backup server 105 to the service being performed.

Here, the HTTP response time is the time from when the user requests the desired service to the completion of the screen display, and the FTP response time is the time when the user requests the desired service and the file transfer is completed. It is time to do. The thresholds are determined corresponding to the HTTP response time and the FTP response time, respectively.
The above service level monitoring tool measures these response times, and detects a service whose service level is decreasing by comparing the measured level with a threshold value.

Each of the load balancers 107 and 108 has a W
This device is provided for the eb server and the Ftp server and distributes the load to the service A and the service B. These load balancers reinforce servers by allocating backup servers 105 to overloaded services based on commands from service level monitoring tools on servers 106, and distribute access to services from networks. Things. The network 109 is, for example, an Internet network.

Next, a specific operation of the present embodiment will be described. FIG. 2 is a flowchart for explaining the operation of the present embodiment. This will be described in conjunction with FIG. First, it is assumed that a load is applied from the user in step 101, that is, for example, the user is accessing the server 101 or the server 102 from the network 109 via the load distribution device 107 in order to receive the service A. . Alternatively, since the user receives the service B, the load balancer 108
It is assumed that the server 103 or the server 104 is accessed via the.

In the state where the load is applied by the user, the service level monitoring tool on the server 106 executes the server 101, 102, 103,
The service level is measured for 104 and 105 (step 102). That is, the Internet service level of response time such as HTTP or FTP is measured for each server.

When the measurement of the service level is completed, the measured service level is compared with a preset threshold value (step 103). When the measured level is lower than the threshold value, it is determined that the service level is satisfied. Judge and proceed to step 104 to continue the measurement as it is.
The same measurement is performed after a certain time in Step 2. Since the service level is satisfied when the measurement level is equal to or lower than the threshold value, the processing of steps 102 to 104 is repeated to measure the service level periodically.

On the other hand, when the measured level exceeds the threshold value in step 103, it is determined that the service level is not satisfied, and the service level monitoring tool on the server 106 executes a process for recovering the service level. . For example, if the service level of the service A is not satisfied, a process of reinforcing the server with respect to the service A is performed (step 105). In other words, the server 106 sets the backup server 105 to the We
Activate with service b (step 106).

The server 106 is a load balancer 107
Requesting that the backup server 105 be added as a server that executes processing relating to the service A, and in response to this request, the load distribution apparatus 107 adds the server 105 to the current servers 101 and 102 for the service A (step 107). . As a result, with respect to the service A, processing is performed by distributing the load by a total of three servers, namely, the servers 101, 102, and 105 (step 1).
08) Compensate for the drop in service level for service A.

On the other hand, when the service level of the service B is not satisfied, the same processing as in the case of the service A is performed. That is, the server 106 is the backup server 10
In order to construct an environment capable of providing the service B with respect to No. 5, the backup server 105 is started by the Ftp service (step 106).

The server 106 is connected to a load balancer 108
Requesting that the backup server 105 be added as a server that executes processing relating to the service B, and in response to this request, the load distribution apparatus 108 adds the server 105 to the current servers 103 and 104 for the service B (step 107). . As a result, for the service B, processing is performed by distributing the load by a total of three servers 103, 104, and 105 (step 1).
08) As in the case of the service A, the service level of the service B is compensated for.

As described above, in the present embodiment, the service level is periodically measured for a plurality of services, and when the measured level falls below a predetermined threshold value, the load is distributed by allocating a backup server. Therefore, it is possible to reinforce the server in a timely manner for services whose service level is decreasing. Therefore, even if the load is concentrated on a specific service, the load can be distributed in order to reinforce the server, and for example, it is possible to prevent a decrease in service such as a delay in response to access from a user.

FIG. 3 is a block diagram showing the configuration of the second embodiment of the present invention. In the second embodiment, the server 106 having the service level monitoring tool
The servers are connected to the respective servers via 7, 108. Other configurations are the same as those of the first embodiment. In the present embodiment, the service level monitoring tool on the server 106 periodically measures the Internet service level such as the response time of HTTP, FTP, and NNTP, and based on the measurement result, monitors the service whose service level is decreasing. Perform processing to reinforce the server. The method of measuring the service level is the same as in the first embodiment.

FIG. 4 is a flowchart for explaining the operation of this embodiment. The basic operation is the same as in FIG. First, it is assumed that a load is applied from the user in step 201, that is, for example, the user is accessing the server 101 or the server 102 from the network 109 via the load distribution device 107 in order to receive the service A. . Alternatively, the user receives the service B from the network 109 via the load balancer 108 via the server 103 or the server 10.
4 is accessed.

In the state where the load is applied from the user, the service level monitoring tool on the server 106 accesses the load balancers 107 and 108 at regular time intervals and checks the service levels of the service levels A and B. A measurement is performed (step 202). The method for measuring the service level is the same as in the first embodiment, and a description thereof will be omitted.

When the measurement of the service level is completed, the measured service level is compared with a preset threshold (step 203). When the measured level is lower than the threshold, it is determined that the service level is satisfied. Judgment proceeds to step 204 to continue the measurement as it is, step 20
The same measurement is performed after a certain time in Step 2. When the measurement level is equal to or lower than the threshold as described above, the service level is satisfied. Therefore, the processing of steps 202 to 204 is repeated, and the service level is periodically measured.

On the other hand, when the measured level exceeds the threshold value in step 203, it is determined that the service level is not satisfied, and the service level monitoring tool on the server 106 executes a process for recovering the service level. . For example, if the service level of the service A is not satisfied, a process of reinforcing the server with respect to the service A is performed (step 205). That is, the server 106 sets the backup server 105 to a Web environment in order to construct an environment capable of providing the service A with respect to the backup server 105.
Activate with service b (step 206).

The server 106 is provided with a load balancer 107
Requesting that the backup server 105 be added as a server that executes processing relating to the service A, and in response to this request, the load distribution apparatus 107 adds the server 105 to the current servers 101 and 102 for the service A (step 207). . As a result, with respect to the service A, the processing is performed by distributing the load by the total of three servers 101, 102, and 105 (step 2).
08) Compensate for the drop in service level for service A.

On the other hand, when the service level of the service B is not satisfied, the same processing as in the case of the service A is performed. That is, the server 106 is the backup server 10
In order to construct an environment capable of providing the service B with respect to No. 5, the backup server 105 is started by the Ftp service (step 206).

The server 106 is connected to a load distribution device 108
Requesting that the backup server 105 be added as a server that executes processing relating to the service B, and in response to this request, the load balancer 108 adds the server 105 to the current servers 103 and 104 for the service B (step 207). . As a result, for the service B, processing is performed by distributing the load by a total of three servers 103, 104, and 105 (step 2).
08) As in the case of the service A, the service level of the service B is compensated for.

Also in this embodiment, the service level is periodically measured for a plurality of services, and when the measured level falls below a predetermined threshold value, the load is distributed by allocating a backup server. In addition, it is possible to timely reinforce a server for a service whose service level is decreasing. Therefore, even if the load is concentrated on a specific service, the load can be distributed to reinforce the server and the service can be prevented from lowering.

[0028]

As described above, according to the present invention, the service levels of a plurality of services are periodically measured, and a backup server is added to a service whose service level has been reduced, so that a service from the network is added. Since the load is distributed, it is possible to reinforce the server in real time, and even if the load is concentrated on a specific service, the load can be distributed in order to reinforce the server in a timely manner. Can be prevented. In addition, by using a server that operates for any of a plurality of services, it is not necessary to prepare a server for each service, so that the scale of the system does not increase.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a first exemplary embodiment of the present invention.

FIG. 2 is a flowchart for explaining the operation of the embodiment of FIG. 1;

FIG. 3 is a block diagram illustrating a configuration of a second exemplary embodiment of the present invention.

FIG. 4 is a flowchart for explaining the operation of the embodiment of FIG. 3;

[Explanation of symbols]

 101 to 104 server 105 backup server 106 server (service level monitoring tool) 107, 108 load balancer 109 network

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06F 15/00 320 G06F 15/00 320C 320L F-term (Reference) 5B034 BB11 DD02 5B045 GG04 JJ05 JJ08 5B085 AA05 AA08 AC11 AC12 5B089 GA11 HA10 HB04 HB05 JB18 KA05 KA06 KC28 MA05 MA06 MA07

Claims (6)

[Claims] 1. A means for periodically measuring service levels of a plurality of services in a network system in which a server is connected to a network corresponding to each of the plurality of services and provides a plurality of services on the network, Means for comparing the measured level of each service measured by the measuring means with a preset threshold value, and when the measured level exceeds the threshold value based on the comparison result of the comparing means, the measured level is Assign a backup server to the service that exceeds the threshold,
A network system comprising: means for reinforcing a server. 2. The network system according to claim 1, wherein the measuring unit measures a service level by measuring an HTTP response time. 3. The network system according to claim 1, wherein the measuring unit measures a service level by measuring an FTP response time. 4. The network system according to claim 1, wherein the backup server is a server operable for a plurality of services. 5. The network system according to claim 1, wherein said measuring means measures service levels of a plurality of services at predetermined time intervals. 6. The network system according to claim 1, wherein said network is an Internet network.